Experiments in Agentic AI for Science

科学领域代理式人工智能的实验

Abstract: This paper details two novel frameworks for developing autonomous, agentic AI in scientific workflows. Both systems leverage a hybrid Local Body, Remote Brain architecture via Google Colab, utilizing Python-based local orchestrators to invoke large language model (LLM) cloud backends.

摘要： 本文详细介绍了两种用于在科学工作流中开发自主代理式人工智能（Agentic AI）的新型框架。这两个系统均利用了通过 Google Colab 实现的“本地躯干、远程大脑”（Local Body, Remote Brain）混合架构，并使用基于 Python 的本地编排器来调用云端的大型语言模型（LLM）后端。

The first agent, DeepTS/DeepCollector, automates the large-scale curation, extraction, and deduplication of time-series datasets. The second, DeepScribe, is an autonomous presentation analyzer that converts visually dense, mathematically complex physics lectures into structured scientific reports.

第一个代理 DeepTS/DeepCollector 实现了时间序列数据集的大规模整理、提取和去重自动化。第二个代理 DeepScribe 是一个自主演示分析器，它能将视觉信息密集且数学复杂的物理学讲座转换为结构化的科学报告。

Through practical systems engineering—such as granular attribute extraction (Cellular RAG), remote data inspection, and distributed concurrency controls—we demonstrate how agentic AI can overcome the context and reasoning limitations of current state-of-the-art systems to rigorously support scientific workflows.

通过实际的系统工程实践——例如细粒度属性提取（Cellular RAG）、远程数据检查以及分布式并发控制——我们展示了代理式人工智能如何克服当前最先进系统在上下文和推理能力上的局限性，从而为科学工作流提供严谨的支持。

Finally, we outline a generalization of DeepTS to support deep knowledge graphs and discuss the application of this conceptual approach to high-energy physics (DeepQCD).

最后，我们概述了 DeepTS 的泛化方案以支持深度知识图谱，并讨论了这一概念方法在高能物理领域（DeepQCD）的应用。